Every incident follows a timestamped cycle — investigating → identified → monitoring → resolved — each step attributed to the team member who performed it. Nothing gets communicated without a trace.
Multi-region monitoring triggers the incident, or a user reports it via "Report an issue".
4 timestamped steps, each attributed to a team member — a reviewable history, not just a status.
Real-time status page, email subscription, RSS feed, and maintenance windows announced ahead of time.
Anti-spam cooldown and daily summaries — the alert gets through without flooding the team.
Email, Slack, Discord, Teams, Telegram, PagerDuty, Splunk, Google Chat, Ntfy.
Declare an incident yourself, even when automated monitoring didn't see it coming.